EST architecture with LCA and CSA achieves fully unified sequence modeling, significantly improving industrial CTR prediction performance.

Mingyang Liu, Yong Bai, Zhangming Chan et al.

UMEM framework improves multi-turn task performance by 10.67% through unified memory extraction and management.

Yongshi Ye, Hui Jiang, Feihu Jiang et al.

BNRM integrates Bayesian non-negative factor analysis to mitigate reward hacking in RLHF, improving robustness and interpretability.

Zhibin Duan, Guowei Rong, Zhuo Li et al.

Introduces GRU-Mem for long-context reasoning, achieving up to 400% efficiency improvement by addressing memory explosion and lack of exit mechanisms.

Leheng Sheng, Yongtao Zhang, Wenchang Ma et al.

LAP enables zero-shot cross-embodiment robot control by representing low-level actions as natural language, achieving over 50% success rate.

Lihan Zha, Asher J. Hancock, Mingtong Zhang et al.

C2RoPE introduces spatio-temporal continuous positional encoding and Chebyshev causal masking, improving 3D multimodal reasoning by addressing RoPE limitations.

Guanting Ye, Qiyan Zhao, Wenhao Yu et al.

DAWN method improves value learning efficiency in residual RL through data anchoring and normalization.

Guozheng Ma, Lu Li, Haoyu Wang et al.

RLTT distributes rewards across latent reasoning steps, significantly boosting LoopLM's performance on math and reasoning benchmarks.

Jonathan Williams, Esin Tureci

Proposes hierarchical goal planning and simulation validation in AIvilization v0 for large-scale social simulation with long-term autonomy.

Wenkai Fan, Shurui Zhang, Xiaolong Wang et al.

MLDocRAG improves retrieval accuracy and generation coherence through Multimodal Chunk-Query Graph.

Yongyue Zhang, Yaxiong Wu

Proposes Separable Causal Diffusion (SCD) architecture, revealing early layers' feature redundancy and deep layers' local focus, boosting video generation efficiency.

Xingjian Bai, Guande He, Zhengqi Li et al.

UniVTAC creates a simulation platform supporting three tactile sensors, training a multi-task encoder that improves tactile perception by 17.1% on benchmarks and 25% in real robots.

Baijun Chen, Weijie Wan, Tianxing Chen et al.

Introduced Chain of Mindset framework, improving accuracy by 4.96% on Qwen3-VL-32B-Instruct.

Tianyi Jiang, Arctanx An, Hengyi Feng et al.

WildCat uses randomly pivoted Cholesky to select a small weighted coreset for near-linear time attention.

Tobias Schröder, Lester Mackey

Proposes ScaleSWE, an automated multi-agent system for large-scale construction of high-quality SWE datasets, producing 100k verified instances.

Jiale Zhao, Guoxin Chen, Fanzhe Meng et al.

Introduces Toeplitz-based spectral methods for data-driven dynamical systems feature estimation.

Vladimir R. Kostic, Karim Lounici, Massimiliano Pontil

Proposes AI capabilities and propensities as causal dispositional properties; critiques benchmark and elicitation methods; advocates for causal, hypothesis-driven measurement.

Konstantinos Voudouris, Mirko Thalmann, Alex Kipnis et al.

Proposes ExO-PPO, combining conservative on-policy guarantees with off-policy data reuse to enhance sample efficiency and stability.

Hanyong Wang, Menglong Yang

KVCodec leverages GPU-native video codecs to compress remote KV caches, reducing TTFT by up to 3.51× while maintaining accuracy.

Liang Mi, Weijun Wang, Jinghan Chen et al.

LEMUR creates a multilingual EU law corpus and fine-tunes state-of-the-art embedding models with contrastive learning to improve legal retrieval accuracy.

Narges Baba Ahmadi, Jan Strich, Martin Semmann et al.